Face Generation¶
In this project, you'll define and train a DCGAN on a dataset of faces. Your goal is to get a generator network to generate new images of faces that look as realistic as possible!
The project will be broken down into a series of tasks from loading in data to defining and training adversarial networks. At the end of the notebook, you'll be able to visualize the results of your trained Generator to see how it performs; your generated samples should look like fairly realistic faces with small amounts of noise.
Get the Data¶
You'll be using the CelebFaces Attributes Dataset (CelebA) to train your adversarial networks.
This dataset is more complex than the number datasets (like MNIST or SVHN) you've been working with, and so, you should prepare to define deeper networks and train them for a longer time to get good results. It is suggested that you utilize a GPU for training.
Pre-processed Data¶
Since the project's main focus is on building the GANs, we've done some of the pre-processing for you. Each of the CelebA images has been cropped to remove parts of the image that don't include a face, then resized down to 64x64x3 NumPy images. Some sample data is show below.
If you are working locally, you can download this data by clicking here
This is a zip file that you'll need to extract in the home directory of this notebook for further loading and processing. After extracting the data, you should be left with a directory of data processed_celeba_small/
data_dir = 'images/'
"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
import pickle as pkl
import matplotlib.pyplot as plt
import numpy as np
import problem_unittests as tests
#import helper
%matplotlib inline
Visualize the CelebA Data¶
The CelebA dataset contains over 200,000 celebrity images with annotations. Since you're going to be generating faces, you won't need the annotations, you'll only need the images. Note that these are color images with 3 color channels (RGB)#RGB_Images) each.
Pre-process and Load the Data¶
Since the project's main focus is on building the GANs, we've done some of the pre-processing for you. Each of the CelebA images has been cropped to remove parts of the image that don't include a face, then resized down to 64x64x3 NumPy images. This pre-processed dataset is a smaller subset of the very large CelebA data.
There are a few other steps that you'll need to transform this data and create a DataLoader.
Exercise: Complete the following get_dataloader function, such that it satisfies these requirements:¶
- Your images should be square, Tensor images of size
image_size x image_sizein the x and y dimension. - Your function should return a DataLoader that shuffles and batches these Tensor images.
ImageFolder¶
To create a dataset given a directory of images, it's recommended that you use PyTorch's ImageFolder wrapper, with a root directory processed_celeba_small/ and data transformation passed in.
Data Citation¶
@inproceedings{liu2015faceattributes, title = {Deep Learning Face Attributes in the Wild}, author = {Liu, Ziwei and Luo, Ping and Wang, Xiaogang and Tang, Xiaoou}, booktitle = {Proceedings of International Conference on Computer Vision (ICCV)}, month = {December}, year = {2015} }
# necessary imports
import torch
from torchvision import datasets
from torchvision import transforms
from torch.utils.data import DataLoader
def get_dataloader(batch_size, image_size, data_dir='images/'):
"""
Batch the neural network data using DataLoader
:param batch_size: The size of each batch; the number of images in a batch
:param img_size: The square size of the image data (x, y)
:param data_dir: Directory where image data is located
:return: DataLoader with batched data
"""
# TODO: Implement function and return a dataloader
transformation = transforms.Compose([transforms.Resize(image_size),
transforms.CenterCrop(image_size),
transforms.ToTensor()])
data = datasets.ImageFolder(data_dir,transform=transformation)
loader = DataLoader(data,batch_size=batch_size,shuffle= True)
return loader
Create a DataLoader¶
Exercise: Create a DataLoader celeba_train_loader with appropriate hyperparameters.¶
Call the above function and create a dataloader to view images.
- You can decide on any reasonable
batch_sizeparameter - Your
image_sizemust be32. Resizing the data to a smaller size will make for faster training, while still creating convincing images of faces!
# Define function hyperparameters
batch_size = 32
img_size = 32
"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
# Call your function and get a dataloader
celeba_train_loader = get_dataloader(batch_size, img_size)
Next, you can view some images! You should seen square images of somewhat-centered faces.
Note: You'll need to convert the Tensor images into a NumPy type and transpose the dimensions to correctly display an image, suggested imshow code is below, but it may not be perfect.
# helper display function
def imshow(img):
npimg = img.numpy()
plt.imshow(np.transpose(npimg, (1, 2, 0)))
"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
# obtain one batch of training images
dataiter = iter(celeba_train_loader)
images, _ = dataiter.next() # _ for no labels
# plot the images in the batch, along with the corresponding labels
fig = plt.figure(figsize=(20, 4))
plot_size=20
for idx in np.arange(plot_size):
ax = fig.add_subplot(2, plot_size/2, idx+1, xticks=[], yticks=[])
imshow(images[idx])
Exercise: Pre-process your image data and scale it to a pixel range of -1 to 1¶
You need to do a bit of pre-processing; you know that the output of a tanh activated generator will contain pixel values in a range from -1 to 1, and so, we need to rescale our training images to a range of -1 to 1. (Right now, they are in a range from 0-1.)
# TODO: Complete the scale function
def scale(x, feature_range=(-1, 1)):
''' Scale takes in an image x and returns that image, scaled
with a feature_range of pixel values from -1 to 1.
This function assumes that the input x is already scaled from 0-1.'''
# assume x is scaled to (0, 1)
# scale to feature_range and return scaled x
min,max = feature_range
x = x * (max-min) + min
return x
"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
# check scaled range
# should be close to -1 to 1
img = images[0]
scaled_img = scale(img)
print('Min: ', scaled_img.min())
print('Max: ', scaled_img.max())
Define the Model¶
A GAN is comprised of two adversarial networks, a discriminator and a generator.
Discriminator¶
Your first task will be to define the discriminator. This is a convolutional classifier like you've built before, only without any maxpooling layers. To deal with this complex data, it's suggested you use a deep network with normalization. You are also allowed to create any helper functions that may be useful.
Exercise: Complete the Discriminator class¶
- The inputs to the discriminator are 32x32x3 tensor images
- The output should be a single value that will indicate whether a given image is real or fake
import torch.nn as nn
import torch.nn.functional as F
from torch.nn.utils import spectral_norm
class Discriminator(nn.Module):
def __init__(self, conv_dim):
"""
Initialize the Discriminator Module
:param conv_dim: The depth of the first convolutional layer
"""
super(Discriminator, self).__init__()
# complete init function
self.conv_dim = conv_dim
#define layers
#first conv layer reducing dimensionality by 2 and increasing conv_dim to conv_dim
self.conv1 = nn.Conv2d(3,conv_dim,4,padding=1,stride=2)
#second conv layer reducing dimensionality by 2 and increasing conv_dim to 2
self.conv2 = nn.Conv2d(conv_dim , conv_dim*2,4,padding=1,stride=2)
#instance norm after 2nd conv layer to increase model performance
self.in2 = nn.InstanceNorm2d(conv_dim*2)
#third conv layer reducing dimensionality by 2 and increasing conv_dim to 4
self.conv3 = nn.Conv2d(conv_dim*2,conv_dim*4,4,padding=1,stride=2)
#instance norm after 3rd conv layer to increase model performance
self.in3 = nn.InstanceNorm2d(conv_dim*4)
#final linear layer with 1 output, BCEloss will be calculated so no need to apply sigmoid.
self.fcout = nn.Linear(conv_dim*8*8,1)
#1d and 2d dropout layers to improve results.
self.dropout = nn.Dropout2d(p=0.3)
self.dropout1d = nn.Dropout(p=0.3)
def forward(self, x):
"""
Forward propagation of the neural network
:param x: The input to the neural network
:return: Discriminator logits; the output of the neural network
"""
# define feedforward behavior
#pass input tensor thru network
x = F.leaky_relu(self.conv1(x))
x = self.dropout(x)
x = self.conv2(x)
x = F.leaky_relu(self.in2(x))
x = self.dropout(x)
x = self.conv3(x)
x = F.leaky_relu(self.in3(x))
x = self.dropout(x)
#reshape vector to fit into linear layer
x = x.view(-1,self.conv_dim*8*8)
x = self.dropout1d(x)
out = self.fcout(x)
#return logit of linear layer
return out
"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_discriminator(Discriminator)
Generator¶
The generator should upsample an input and generate a new image of the same size as our training data 32x32x3. This should be mostly transpose convolutional layers with normalization applied to the outputs.
Exercise: Complete the Generator class¶
- The inputs to the generator are vectors of some length
z_size - The output should be a image of shape
32x32x3
class Generator(nn.Module):
def __init__(self, z_size, conv_dim):
"""
Initialize the Generator Module
:param z_size: The length of the input latent vector, z
:param conv_dim: The depth of the inputs to the *last* transpose convolutional layer
"""
super(Generator, self).__init__()
# complete init function
self.z_size = z_size
self.conv_dim = conv_dim
#define layers
#dense layer reshapping fron z_size to conv_dim*16
self.fcin = nn.Linear(z_size,conv_dim*16)
#first conv tranpose layer increasing dimensionality by 2 and decraising conv_dim to 8
self.deconv1 = nn.ConvTranspose2d(conv_dim*16,conv_dim*8,4,padding=1,stride=2)
#instance norm after 1st deconv layer to increase model performance
self.in1 = nn.InstanceNorm2d(conv_dim*8)
#second convtranspose layer increasing dimensionality by 2 and decreasing conv_dim to 4
self.deconv2 = nn.ConvTranspose2d(conv_dim*8,conv_dim*4,4,padding=1,stride=2)
#instance norm after 2nd deconv layer to increase model performance
self.in2 = nn.InstanceNorm2d(conv_dim*4)
#third conv transpose layer increasing dimensionality by 2 and decreasing conv_dim to 2
self.deconv3 = nn.ConvTranspose2d(conv_dim*4,conv_dim*2,4,padding=1,stride=2)
#instance norm after 3rd deconv layer to increase model performance
self.in3 = nn.InstanceNorm2d(conv_dim*2)
#fourth conv transpose layer increasing dimensionality by 2 and decreasing conv_dim to conv_dim
self.deconv4 = nn.ConvTranspose2d(conv_dim*2 , conv_dim,4,padding=1,stride=2)
#instance norm after 4th conv layer to increase model performance
self.in4 = nn.InstanceNorm2d(conv_dim)
#last conv transpose layer increasing dimensionality by 2 and increasing conv_dim to 3
self.deconv5 = nn.ConvTranspose2d(conv_dim,3,4,padding=1,stride=2)
#put in dropout layers to improve results
self.dropout1d = nn.Dropout(p=0.3)
self.dropout2d = nn.Dropout2d(p=0.3)
def forward(self, x):
"""
Forward propagation of the neural network
:param x: The input to the neural network
:return: A 32x32x3 Tensor image as output
"""
# define feedforward behavior
#pass input thru first linear layer
x = F.relu(self.fcin(x))
#reshape to to 4d tensor to fit into conv layers
x = x.view(-1,self.conv_dim*16,1,1)
#pass thru conv layers
x = self.dropout1d(x)
x = F.relu(self.deconv1(x))
x = self.dropout2d(x)
x = self.deconv2(x)
x = F.relu(self.in2(x))
x = self.dropout2d(x)
x = self.deconv3(x)
x = F.relu(self.in3(x))
x = self.dropout2d(x)
x = self.deconv4(x)
x = F.relu(self.in4(x))
x = self.dropout2d(x)
#pass final output thru tan activation function
out = F.tanh(self.deconv5(x))
return out
"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_generator(Generator)
Initialize the weights of your networks¶
To help your models converge, you should initialize the weights of the convolutional and linear layers in your model. From reading the original DCGAN paper, they say:
All weights were initialized from a zero-centered Normal distribution with standard deviation 0.02.
So, your next task will be to define a weight initialization function that does just this!
You can refer back to the lesson on weight initialization or even consult existing model code, such as that from the networks.py file in CycleGAN Github repository to help you complete this function.
Exercise: Complete the weight initialization function¶
- This should initialize only convolutional and linear layers
- Initialize the weights to a normal distribution, centered around 0, with a standard deviation of 0.02.
- The bias terms, if they exist, may be left alone or set to 0.
def weights_init_normal(m):
"""
Applies initial weights to certain layers in a model .
The weights are taken from a normal distribution
with mean = 0, std dev = 0.02.
:param m: A module or layer in a network
"""
# classname will be something like:
# `Conv`, `BatchNorm2d`, `Linear`, etc.
classname = m.__class__.__name__
# TODO: Apply initial weights to convolutional and linear layers
#class is a conv layer or linear layer, adjust weights
if 'Conv' in classname or 'Linear' in classname:
m.weight.data.uniform_(0.0, 0.02)
m.bias.data.fill_(0)
Build complete network¶
Define your models' hyperparameters and instantiate the discriminator and generator from the classes defined above. Make sure you've passed in the correct input arguments.
"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
def build_network(d_conv_dim, g_conv_dim, z_size):
# define discriminator and generator
D = Discriminator(d_conv_dim)
G = Generator(z_size=z_size, conv_dim=g_conv_dim)
# initialize model weights
D.apply(weights_init_normal)
G.apply(weights_init_normal)
print(D)
print()
print(G)
return D, G
Exercise: Define model hyperparameters¶
# Define model hyperparams
d_conv_dim = 128
g_conv_dim = 128
z_size = 100
"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
D, G = build_network(d_conv_dim, g_conv_dim, z_size)
Training on GPU¶
Check if you can train on GPU. Here, we'll set this as a boolean variable train_on_gpu. Later, you'll be responsible for making sure that
- Models,
- Model inputs, and
- Loss function arguments
Are moved to GPU, where appropriate.
"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
import torch
# Check for a GPU
train_on_gpu = torch.cuda.is_available()
if not train_on_gpu:
print('No GPU found. Please use a GPU to train your neural network.')
else:
print('Training on GPU!')
Discriminator and Generator Losses¶
Now we need to calculate the losses for both types of adversarial networks.
Discriminator Losses¶
- For the discriminator, the total loss is the sum of the losses for real and fake images,
d_loss = d_real_loss + d_fake_loss.- Remember that we want the discriminator to output 1 for real images and 0 for fake images, so we need to set up the losses to reflect that.
Generator Loss¶
The generator loss will look similar only with flipped labels. The generator's goal is to get the discriminator to think its generated images are real.
Exercise: Complete real and fake loss functions¶
You may choose to use either cross entropy or a least squares error loss to complete the following real_loss and fake_loss functions.
def real_loss(D_out, smooth=False):
# compare logits to real labels
# smooth labels if smooth=True
#get batch size of output
batch_size = D_out.size(0)
if smooth:
#create row of torch 1's as pos labels
labels = torch.ones(batch_size) * 0.9
else:
#create row of torch 1's as pos labels
labels = torch.ones(batch_size)
#define loss method
labels = labels.cuda()
criterion = nn.BCEWithLogitsLoss()
#return loss, squeezing out any empty dims
return criterion(D_out.squeeze(),labels.squeeze())
def fake_loss(D_out):
# compare logits to fake labels
#get batch size
batch_size = D_out.size(0)
#make row of zeros as neg label
labels = torch.zeros(batch_size)
labels = labels.cuda()
#define loss method
criterion = nn.BCEWithLogitsLoss()
#return loss, squeezing out any empty dims
return criterion(D_out.squeeze(),labels.squeeze())
import torch.optim as optim
# Create optimizers for the discriminator D and generator G
#use default lr for Adam
d_optimizer = optim.Adam(D.parameters(),lr= 0.0005 ,betas=[0.3,0.999])
g_optimizer = optim.Adam(G.parameters(),lr = 0.0005 , betas=[0.3,0.999])
def view_samples(epoch, samples):
fig, axes = plt.subplots(figsize=(16,4), nrows=2, ncols=8, sharey=True, sharex=True)
for ax, img in zip(axes.flatten(), samples[epoch]):
img = img.detach().cpu().numpy()
img = np.transpose(img, (1, 2, 0))
img = ((img + 1)*255 / (2)).astype(np.uint8)
ax.xaxis.set_visible(False)
ax.yaxis.set_visible(False)
im = ax.imshow(img.reshape((32,32,3)))
Training¶
Training will involve alternating between training the discriminator and the generator. You'll use your functions real_loss and fake_loss to help you calculate the discriminator losses.
- You should train the discriminator by alternating on real and fake images
- Then the generator, which tries to trick the discriminator and should have an opposing loss function
Saving Samples¶
You've been given some code to print out some loss statistics and save some generated "fake" samples.
Exercise: Complete the training function¶
Keep in mind that, if you've moved your models to GPU, you'll also have to move any model inputs to GPU.
def train(D, G, n_epochs, print_every=50):
'''Trains adversarial networks for some number of epochs
param, D: the discriminator network
param, G: the generator network
param, n_epochs: number of epochs to train for
param, print_every: when to print and record the models' losses
return: D and G losses'''
# move models to GPU
if train_on_gpu:
D.cuda()
G.cuda()
# keep track of loss and generated, "fake" samples
samples = []
losses = []
# Get some fixed data for sampling. These are images that are held
# constant throughout training, and allow us to inspect the model's performance
sample_size=16
fixed_z = np.random.uniform(-1, 1, size=(sample_size, z_size))
fixed_z = torch.from_numpy(fixed_z).float()
# move z to GPU if available
if train_on_gpu:
fixed_z = fixed_z.cuda()
# epoch training loop
for epoch in range(n_epochs):
# batch training loop
for batch_i, (real_images, _) in enumerate(celeba_train_loader):
batch_size = real_images.size(0)
real_images = scale(real_images).cuda()
# ===============================================
# YOUR CODE HERE: TRAIN THE NETWORKS
# ===============================================
# 1. Train the discriminator on real and fake images
#clear optimizer gradients
d_optimizer.zero_grad()
# Generate latent vector
z = np.random.uniform(-1, 1, size=(batch_size, z_size))
z = torch.from_numpy(z).float().cuda()
#generate fake img
fake_img = G(z)
#put real image thru discrimator
real_img_d= D(real_images)
#Put fake image thru discrimator
fake_img_d = D(fake_img)
#generate real loss with smoothing
real_l = real_loss(real_img_d,smooth=True)
#generate fake loss
fake_l = fake_loss(fake_img_d)
#sum up losses
d_loss = real_l + fake_l
#backprop thru network
d_loss.backward()
#take optimization step
d_optimizer.step()
# 2. Train the generator with an adversarial loss
g_optimizer.zero_grad()
# Generate latent vector
z = np.random.uniform(-1, 1, size=(batch_size, z_size))
z = torch.from_numpy(z).float().cuda()
#generate fake img
fake_img = G(z)
#put fake image thru discrimanator
fake_img_d = D(fake_img)
#generate real loss on fake_img
real_l = real_loss(fake_img_d,smooth=True)
#sum up losses
g_loss = real_l
#backprop thru network
g_loss.backward()
#take optimization step
g_optimizer.step()
# ===============================================
# END OF YOUR CODE
# ===============================================
# Print some loss stats
if batch_i % print_every == 0:
# append discriminator loss and generator loss
losses.append((d_loss.item(), g_loss.item()))
# print discriminator and generator loss
print('Epoch [{:5d}/{:5d}] | d_loss: {:6.4f} | g_loss: {:6.4f}'.format(
epoch+1, n_epochs, d_loss.item(), g_loss.item()))
## AFTER EACH EPOCH##
# this code assumes your generator is named G, feel free to change the name
# generate and save sample, fake images
G.eval() # for generating samples
samples_z = G(fixed_z)
samples.append(samples_z)
G.train() # back to training mode
#view images
view_samples(epoch,samples)
# Save training generator samples
with open('train_samples.pkl', 'wb') as f:
pkl.dump(samples, f)
# finally return losses
return losses
Set your number of training epochs and train your GAN!
# set number of epochs
n_epochs = 25
"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
# call training function
losses = train(D, G, n_epochs=n_epochs)
Training loss¶
Plot the training losses for the generator and discriminator, recorded after each epoch.
fig, ax = plt.subplots()
losses = np.array(losses)
plt.plot(losses.T[0], label='Discriminator', alpha=0.5)
plt.plot(losses.T[1], label='Generator', alpha=0.5)
plt.title("Training Losses")
plt.legend()
Generator samples from training¶
View samples of images from the generator, and answer a question about the strengths and weaknesses of your trained models.
# helper function for viewing a list of passed in sample images
import pickle as pkl
import matplotlib.pyplot as plt
import numpy as np
def view_samples(epoch, samples):
fig, axes = plt.subplots(figsize=(16,4), nrows=2, ncols=8, sharey=True, sharex=True)
for ax, img in zip(axes.flatten(), samples[epoch]):
img = img.detach().cpu().numpy()
img = np.transpose(img, (1, 2, 0))
img = ((img + 1)*255 / (2)).astype(np.uint8)
ax.xaxis.set_visible(False)
ax.yaxis.set_visible(False)
im = ax.imshow(img.reshape((32,32,3)))
# Load samples from generator, taken while training
with open('train_samples.pkl', 'rb') as f:
samples = pkl.load(f)
_ = view_samples(-1, samples)
Question: What do you notice about your generated samples and how might you improve this model?¶
When you answer this question, consider the following factors:
- The dataset is biased; it is made of "celebrity" faces that are mostly white
- Model size; larger models have the opportunity to learn more features in a data feature space
- Optimization strategy; optimizers and number of epochs affect your final result
-My generated samples are all white, which caused at a data in the bias , making most of my generated set white aswell.¶
-My model was large, but perhaps not large enough to capture all the nuances of a human face.¶
-Perhaps a higher learning rate would have resulted in better generated faces.¶
Answer: (Write your answer in this cell)
Submitting This Project¶
When submitting this project, make sure to run all the cells before saving the notebook. Save the notebook file as "dlnd_face_generation.ipynb" and save it as a HTML file under "File" -> "Download as". Include the "problem_unittests.py" files in your submission.